AI image generators don't "draw"—they follow a compass: the score function, which points toward more probable images. The same compass drives Bayesian sampling and plasma physics.
We built DiScoFormer to estimate the score far better when data gets complex. 🧵
Recovering a distribution means estimating its density (where data clusters) + the score.
Kernel density estimation (KDE) does both, but when data has many dimensions, it breaks down. Neural score-matching models stay more accurate but need to be retrained per dataset.
DiScoFormer is a single model that gives both density & score on any dataset with no retraining.
With 100 numbers per point, it cuts error vs. the best-tuned KDE ~6.5× on score and 37× on density.
DiScoFormer can also handle data unlike anything it trained on—more clusters than it ever saw, and unfamiliar shapes.
Score and density estimation normally means training a model per dataset. DiScoFormer does both in one pretrained pass, cutting error roughly 6.5x on score and 37x on density against best-tuned kernel density estimation at 100 dimensions.
Read the full thread on X ↗
